Jump to content

Sparepart

RHOCaR Member
  • Posts

    140
  • Joined

  • Last visited

  • Days Won

    3

Sparepart last won the day on June 4

Sparepart had the most liked content!

Previous Fields

  • Car type
    Robin Hood EXMO
  • Full name
    Not Available

Profile Information

  • Gender
    Male
  • Location
    Chichester

Recent Profile Visitors

1,089 profile views

Sparepart's Achievements

Wheely good builder!

Wheely good builder! (4/5)

13

Reputation

  1. At great expense, I have comissioned a technical drawing of the basic way a fuel gauge is wired up. The principles might help you figure out which wires you need to connect to what in your case.
  2. I agree. I looked at the Kitspares product that you probably have. https://kitspares.co.uk/collections/fuel-system/products/fuel-sender-unit It does not match the drawing above because it is probably for a really basic Sierra maodel, maybe the pickup. The sender with the three pins is more complicated because as well as the variable resistance for the fuel gauge it can also be linked up to two other features of higher spec Sierras, that is, the trip computer and the low fuel level warning lamp. So for most folk with a kit car only one of the three connectors on the sender unit is used. Essentially that is what you have, a sender unit that has the single needed connector.
  3. There is a previous thread that discusses the resistance values at the pins of a working sender. https://www.rhocar.org/index.php?/forums/topic/49494-sender-unit Unfortunatey niducan's photos no longer seem to be accessible, note one pin goes up in resistance, one down and the other stays constant, so it is possible to wire the conections up so that a full tank might read empty and vice versa, I think.
  4. Sparepart

    New 2B Owner

    Am I correct in supposing that the windscreen is removed before taking it for an MOT ? then replaced after ?
  5. Wow! what a car, surrounded by the bits and pieces of my rebuilding, looking at all the work still to do, I heard a whisper saying, "Throw this load of tat away and buy that car!". At least, as I continue, I can see what is possible, eventually, perhaps, sometime, maybe.
  6. I notice that you are a recently joined member, so i feel I have to mention that unless you are replacing an existing hood, and hence have an accurate pattern, there is going to be a fitting exercise involved. This is because during construction the windscreen position was determined by the builder, there were no pre-drilled holes for the pillar mounts in the factory bodyshell. This also applies to any frame or studs that were fitted. That is to say eatch hood will be unique, starting from a hood of generally correct proportions. You also did not mention sidescreens etc ... do you have some already ? I apologise if you are already on top of this stuff, I know that it does not help to get a response that does not suggest a solution, only raises more questions. Although not urgent, I too am on the lookout for a hood, frame, sidescreens etc for an EXMO so will be keeping an eye on this thread.
  7. Ah, I was wondering how it might work, floatation seemed unlikely, I thought perhaps a change in capacity between two plates, but a rise in temp is a good possibility. I threw away the "Auxiliary Warning Module" to which the sender should be connected a long long time ago. I'll check the resistance, if its a heating element there should be a lowish resitance, maybe like 10 ohms. Perhaps then I can gently heat it from outside, wave a flame at it, and see if it goes open circuit. In any event if I do want to use it again, with no Auxiliary Warning Module, I would need to design a circuit to power it correctly and illuminate a lamp when the circuite opens .... mmmm... probably leave that to a winter project.
  8. Her is a photo of one thats for sale on the web.
  9. The hole in the block is about 1/2" diameter on the offside near the mounting arm but right down just above where the sump starts. The sensor is then quite long about 6 inches roughly, a long thin metal probe with a wide "bung" at the top where the sensor wire exits to a connector. When the "bung" is in the hole the end of the sensor is then down in the sump at a similar depth to the end of the dipstick.
  10. During the original (rushed) build I removed the sensor for low engine oil level from the block and tapped a wooden round peg in the round hole. Now, during a rebuild I have more time to consider... should I properly seal the hole, which kept weeping oil .. or reinstate the oil level warning sensor. To be honest I am thinking that it is not needed and just one more thing to leak or go wrong and I am not sure if it is just a single point sensor or a gradual sensor. So before I throw it away I was wondering if anyone else has fitted/used this sensor and has some experience to share.
  11. To reply to your original first question number 5 (yes you have two question no 5s), You want to know how much of the car has to be non standard before various tests are required. This is all explained at the .gov website, look down the list of degrees of modifications, from rebuilt vehicles to radically altered vehicles and see whats needed to get them on the road. Not all need a type approval test. https://www.gov.uk/vehicle-registration
  12. Sparepart

    Fuel line

    On the similarly bodied Exmo I have used the small "microbore" copper pipe (6mm I think probably what you have there). Like Harry it is routed up into the "boot" and the out through the nearside wheel arch to run in the box section under the passenger's arm and to the fuel pump. You can create a "flare" by soldering an olive at each end so that the rubber connectors can be pushed over and a clip put behind the olive. In that long run down the side, I placed a plastic (PVC?) 15mm water pipe. The cooper pipe runs through this to protect it from any sharp bits of SS that have not noticed or can't get to. Long time ago. Only an SVA for me, all the checked on was that it was secured at regular intervals, they dis not measure, just a visual assessment.
  13. I think richyb66 means that the alternator is earthed through the mounting bracket, so it has to be firmly mounted for the belt to drive it, hence should be a good earth otherwise it would fall off. This assumes though that the engine block is earthed. I think that the alternator should give pretty much a constant output voltage which is controlled by the regulator circuitry inside the alternator. The regulator varies the field coil current to keep the output voltage constant. When there is demand for current from a flat battery, or a battery just a bit low after starting, or from light bulb filaments, or cigar lighter socket etc this manifests itself in a drop in resistance at the ouput from the alternator and from school I remember V=IR, so to keep V constant with falling R the I has to get bigger, so the regulator increases the magnetic field coil current in the alternator to generate a bigger current output and the voltage constant. With a bigger magnetic field the alternator gets harder to turn and puts more load on the engine via the dtive belt. So possible drops in voltage could also be caused by a loose drive belt, although you normally hear this squeeling from cold. The regulator is the most likely source of the problem. What happens to the voltage if you turn on the headlights or any other current consuming device ?.
  14. I agree, I have just put mine back together on the bench, outside the car. As the boss is tightened down more thread appears at the nut (obviously), and the spring at the bottom gets compressed as more of the shaft is drwan into the tube. The two thrust washers slide on the shaft, the top one is pushed down by the cam and the bottom one pushed by the top of the spring. When the boss has reached it's final position, firmly pressed down on the taper there is still plenty of compression available at the spring. So the cam is not crushed it only has the spring tension pressing on it. I lifted up the assembled unit and put the bottom end of the shaft on the floor, grasping it by the outer tube (the part that is bolted to the cross member under the scuttle) I can push the outer tube down the shaft against the spring and then it moves back when I stop pushing. There must be a reason for it to be like this, possibly to absorb small compressions that try to push the shaft towards the driver ? reducing/eliminating vibrations ?.
  15. Now I need to look at my column more closely, because my cam is broken, I think that when I put the Mountney boss on the shaft I probably failed to notice the notch that molded into it, so did not put the notch just above the cam, damn!
×
×
  • Create New...